A collaborative study of the EDNAP group regarding Y-chromosome binary polymorphism analysis

Forensic Science International 153 (2005) 103–108 www.elsevier.com/locate/forsciint

A collaborative study of the EDNAP group regarding Y-chromosome binary polymorphism analysis Marı´a Brion a,*, Berit M. Dupuy b, Marielle Heinrich c, Carsten Hohoff c, Bernardette Hoste d, Bertrand Ludes e, Bente Mevag b, Niels Morling f, Harald Niedersta¨tter g, Walther Parson g, Juan Sanchez f, Klaus Bender h, Nathalie Siebert d, Catherine Thacker i, Conceic¸ao Vide j, Angel Carracedo a a

Institute of Legal Medicine, University of Santiago de Compostela, San Francisco s/n, 15782 Santiago de Compostela, Spain b Institute of Legal Medicine, Oslo, Norway c Institute of Legal Medicine, Mu¨nster, Germany d National Institute of Forensic Science, Brussels, Belgium e Institute of Legal Medicine, Strasbourg, France f Department of Forensic Genetics, Institute of Forensic Medicine, University of Copenhagen, Denmark g Institute of Legal Medicine, Innsbruck, Austria h Institute of Legal Medicine, Mainz, Germany i Institute of Cell and Molecular Science, Barts and The London, Queen Mary’s School of Medicine and Dentistry, London, UK j Institute of Legal Medicine, Coimbra, Portugal Available online 15 July 2005

Abstract A collaborative study was carried out by the European DNA Profiling Group (EDNAP) in order to evaluate the performance of Y-chromosome binary polymorphism analysis in different European laboratories. Four blood samples were sent to the laboratories, to be analysed for 11 Y-chromosome single nucleotide polymorphisms (SNPs): SRY-1532, M40, M35, M213, M9, 92R7, M17, P25, M18, M153 and M167. All the labs were also asked to submit a population study including these markers. All participating laboratories reported the same results, indicating the reproducibility and robustness of Y-chromosome SNP typing. A total of 535 samples from six different European populations were also analysed. In Galicia (NW Spain) and Belgium, the most frequent haplogroup was R1b*(xR1b1,R1b3df). Haplogroup F*(xK) is one of the most frequent in Austria and Denmark, while the lowest frequency appear in Belgium. Haplogroup frequencies found in this collaborative study were compared with previously published European Y-chromosome haplogroup data. # 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Y-chromosome; SNPs; DNA; EDNAP

* Corresponding author. Tel.: +34 981582327; fax: +34 981580336. E-mail address: [email protected] (M. Brion). 0379-0738/$ – see front matter # 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.forsciint.2005.06.002

104

M. Brion et al. / Forensic Science International 153 (2005) 103–108

1. Introduction

2. Material and methods

Analysis of Y-chromosome polymorphisms has already become a routine technique in most laboratories involved in forensic testing and kinship analysis. Although Y-STRs are the markers of preference in forensic labs, an increasing interest in Y-SNPs is evident in the field today [1]. SNPs are the simplest and most frequent kind of DNA sequence variation among individuals; their mutation rate is low and they can be analysed in short amplicons using new, high throughput technologies. Whether SNPs will replace STRs as primary method of choice in the forensic labs is a matter of conjecture at present. There is no doubt, however, about the usefulness of SNP typing for some specific purposes. For instance, Ychromosome and mtDNA SNPs are informative in the analysis of geographical origin of a sample, which can be of importance in the investigation of criminal cases and for identification purposes. As a consequence of the interest of several European labs in implementing analysis of Y-chromosome SNPs in routine work, a collaborative exercise was proposed by the European DNA Profiling Group (EDNAP). The first goal was to determine whether uniformity of results of Y-SNP typing could be achieved, independent of the methodology used for genotyping. From the available described markers when the present exercise was planned, a set of 11 Y-chromosome binary polymorphisms was selected, to be typed for two control and two unknown individuals in each participating lab. Each lab was free to decide the methodology used for genotyping, according to what was available to them. Previous to this exercise a preliminary trial was carried out with six Y-chromosome SNPs (M9, SRY-1532, SRY8299, SRY-2627, 92R7 and TAT) with the aim of introducing SNP typing technologies in EDNAP labs. The labs were also asked to perform a population study with the 11 markers. Y-chromosome haplogroup frequencies were reported from six different European populations and comparisons made between these populations and previously reported population studies.

2.1. Samples Four blood stains from laboratory staff (Institute of Legal Medicine, Santiago de Compostela), with informed consent and previous acceptance of the local ethical committee, were sent to the 20 members of the EDNAP group. The samples included two control samples with known Y-SNP haplogroup profiles, and two unknown male samples. In addition, six of the labs submitted population data. A total of 535 individuals were analysed, from Austria (Innsbruck area, 129), Belgium (54), Spain (Galicia, 100), Germany (Mu¨nster area, 95), Denmark (107) and Norway (51). 2.2. Genetic analysis A total of 11 Y-chromosome binary markers were analysed: SRY-1532 (also known as SRY10831), M40 (also known as SRY-8299 or SRY4064), M35, M213, M9, 92R7, M17, P25, M18, M153 and M167 (also known as SRY-2627). The participants were asked to type the samples for the 11 Y-SNPs using any technology that they had available. In order to facilitate the analysis, each group was provided with some information about each of the markers, including sequences of the described amplicons and primers in the literature (data not shown), PCR conditions and the restriction enzymes recognizing the polymorphic sites (Table 1). The intention of the exercise was to see whether the typing results obtained were reproducible and accurate, independent of primers, PCR strategies and analytical techniques used. Haplogroups were named according to the proposal of the Y-Chromosome Consortium [2,3]. 2.3. Statistical analysis Binary marker haplogroup frequencies were estimated by simple gene counting. The Arlequin software version 2.0 [4]

Table 1 Typical PCR conditions and restriction enzymes for SNP typing SNP

PCR conditions

Restriction enzymes

Recognition sites

SRY-1532 M40 M35 M213 M9 92R7 M17 P25 M18 M153 M167

Buffer: 1, dNTPs: 200 mM, MgCl2: 1.5 mM, primers: 0.25 mM each one, Taq: 0.5 U, DNA: 10 ng, H2O: up 25 ml

DraIII BsrBI BmrI MaeII HinfI HindIII

CACACA/GTG GAG/CGG ACTGGGACACA/G A/CGT G/ANTC A/AGCTT

Cac8I MaeIII Tsp509I BsiHKAI

GCN/NGC /GTTAC /AATT GTGCT/C

M. Brion et al. / Forensic Science International 153 (2005) 103–108

105

Table 2 Results of the inter-laboratory exercise Sample

SRY-1532

M40

M35

M213

M9

92R7

M17

P25

Control (1) Control (2) Test (3) Test (4)

G G G G

G G G G

G G G G

C C C C

G C C G

T C C T

+G +G +G +G

C C C C

was used to estimate haplogroup diversity values, to calculate genetic distances as pair-wise values of Fst, and to perform analysis of molecular variance by means of AMOVA. Using the SPSS version 11.0 software package, principal component analysis (PCA) was performed on the haplogroup frequencies detected in the populations investigated and in previously studied populations [5–7].

3. Results 3.1. Inter-comparison exercise Of the 20 member laboratories of the EDNAP group, 8 submitted results (previously 10 labs had sent results for the preliminary exercise with six Y-chromosome SNPs). All the

M18 AA AA AA AA

M153

M167

HG

T T T T

C C C C

P*(xR1ab) F*(xK) F*(xK) P*(xR1ab)

submitted results were correct (Table 2). Several different technologies for genotyping had been used. Most of the labs used minisequencing, or single base extension method using the SNaPshot multiplex Kit (Applied Biosystems). One lab used direct sequencing and another lab used RFLPs and mass spectrometry. Although PCR conditions for the described primers in the literature were provided to each lab, different primers and different PCR conditions (data not shown) were used. In all the cases the results were completely concordant. During the exercise, some labs independently reported double signals or non-consensus results for two markers, P25 and 92R7, and the existence of duplications in these markers was thus demonstrated. As a consequence, three EDNAP labs performed additional research on these duplicated markers [8].

Fig. 1. (Left) Maximum parsimony phylogeny of the Y-chromosome SNPs analysed in the EDNAP exercise. Haplogroups found in the populations studied were represented with different colours. (Right) Relative haplogroup frequencies observed in each population.

1

1

R1b3f R1b3d

2 19 9 1 18 6

31 38 4 55 41 37

R1b1

1 1

1 9 3 2 4

1 3 12 2

13 45 16 32 50 35 2 1

Haplogroup diversity (S.E. values).

54 107 51 100 129 95 Belgium Denmark Norway Galicia Austria Germany

a

1

8 16 9

R1b*(xR1b1,R1b3df) R1a1 R1a*(xR1a1) P*(xR1ab) K*(xP) F*(xK) E3b E*(xE3b) BCD A* n

In the forensic field, the use of inter-laboratory exercises has become a useful tool, especially as a first step of the validation procedure when new markers or technologies need to be implemented. There are several publications describing similar exercises; for example the reproducibility of Y-STR multiplexes, also carried out by the EDNAP group [10], and the quality assurance exercise performed by all the

Population

4. Discussion

Table 3 Haplogroup composition and diversity values observed in the populations

In addition to the inter-lab comparison exercise, all the labs were asked to send population data on the Y-chromosome SNPs used in the study. A total of 535 individuals were analysed distributed among six populations, Austria (Innsbruck area), Belgium, Denmark, Germany (Mu¨nster area), Spain (Galicia) and Norway. Although the set of Y-SNPs studied here allows the detection of 13 possible haplogroups, only 11 of them were observed in our populations (Fig. 1). The haplogroup composition of each population and haplogroup diversities are represented in Table 3. The highest diversity value was found in Norway, while the lowest one was observed in Galicia. As shown in Fig. 1, the haplogroup composition of each population seemed to be quite different. To investigate this heterogeneity, an exact test of sample differentiation was undertaken using the haplogroup frequencies. In most cases significant deviations were found, except for Galicia and Germany (P = 0.086  0.009; 10 000 Markov chain steps) and Galicia and Belgium (P = 0.056  0.008; 10 000 Markov chain steps) showing values near the confidence limit, and Austria and Germany (P = 0.147  0.016). Haplogroup R1b*(xR1b1,R1b3df) was one of the most frequent in most of the population, except Norway. Haplogroup F*(xK) is well represented in all the populations, while R1a1 and E*(xE3b) show more heterogeneous distributions. In order to compare the haplogroup composition detected in our collaborative exercise with previous European studies, a principal component analysis was performed using our data in conjunction with published data of European populations (Fig. 2). A total of 1491 samples distributed among 17 populations were plotted. Austria and Denmark clustered together, slightly apart from Germany, while Belgium and Galicia clustered with France, Italy and the Netherlands, To assess the level of population structure, we estimated various F statistics by means of AMOVA [9]. The Fst value calculated for the entire European sample, comprising six populations without partitioning, was 0.038, indicating that a small proportion of the overall variation resulted from interpopulation differences. When the populations were grouped according to geographical location, in North, Central and South Europe, a low degree of inter-group variability was observed (data not shown).

HgDa

3.2. Population data

0.617 0.664 0.793 0.593 0.719 0.705

(0.060) (0.022) (0.024) (0.035) (0.020) (0.023)

M. Brion et al. / Forensic Science International 153 (2005) 103–108

4 1

106

M. Brion et al. / Forensic Science International 153 (2005) 103–108

107

Fig. 2. Principal component (PC) analysis of haplogroup frequencies of several European populations. The first PC accounts for 72.3% of the variance, whereas the second accounts for 16.9%. Populations studied here are shown as squares and previously studied populations are shown as diamonds.

labs contributing to the Y-chromosome haplotype reference database (YHRD) [11]. These are only two examples, among others available in the literature. The present exercise has demonstrated a clear uniformity in Y-chromosome SNP typing, independent of the strategy used. Among the eight labs collaborating in the control assay, four different technologies were used and all achieved the same results. Six labs chose single base extension using the SNaPshot multiplex kit of Applied Biosystems, some of them as multiplex reactions and others as singleplex reactions. The participating labs were asked to type a small sized population sample, in order to evaluate how informative the selected Y-chromosome SNPs were. Six of the laboratories submitted population data, and because all of them reported correct results in the collaborative exercise, their data was analysed. The population data from a seventh lab were complemented with Y-STR haplotype data and published independently [12]. It is important to avoid ascertainment bias in SNP selection for Y-chromosome studies [3]. In this case, among the 535 samples analysed, a total of 11 haplogroups were detected. Since the number of haplogroups described with the 11 SNPs included in the exercise is 13, it seems that the selected SNPs are appropriate for the populations analysed. When comparisons with previous European population studies were performed, our data showed no significant differences and clusters of neighbouring populations were seen in the PCA plot. The distribution of Y-haplogroups showed significant differences between the populations in the exercise. Although the differences were related to frequency, and not to the presence or absence of particular haplogroups

in specific populations, the results show the potential usefulness of Y-chromosome SNP markers when inferring the geographical origins of samples.

Acknowledgements The technical assistance of Meli Rodrı´guez and Raquel Calvo is highly appreciated. This work is part of the Ph.D. thesis of Marielle Heinrich.

References [1] M. Brion, B. Sobrino, A. Blanco-Verea, M.V. Lareu, A. Carracedo, Hierarchical analysis of 30 Y-chromosome SNPs in European populations, Int. J. Legal Med. 119 (2004) 10–15. [2] Y Chromosome Consortium, A Nomenclature system for the tree of human Y-chromosomal binary haplogroups, Genome Res. 12 (2004) 339–348. [3] M.A. Jobling, C. Tyler-Smith, The human Y chromosome: an evolutionary marker comes of age, Nat. Rev. Genet. 4 (8) (2003) 598–612. [4] S. Schneider, D. Roessli, L. Excoffier, Arlequin ver. 2.000: A Software for Population Genetics Data Analysis, Genetics and Biometry Laboratory, University of Geneva, Switzerland, 2000. [5] O. Semino, G. Passarino, P.J. Oefner, A.A. Lin, S. Arbuzova, L.E. Beckman, G. De Benedictis, P. Francalacci, et al. The genetic legacy of paleolithic Homo sapiens sapiens in extant Europeans: a Y chromosome perspective, Science 290 (2000) 1155–1159. [6] N. Al-Zahery, O. Semino, G. Benuzzi, C. Megri, G. Passarino, A. Torroni, A.S. Santachiara-Benerecetti, Y-chromosome and

108

M. Brion et al. / Forensic Science International 153 (2005) 103–108

mtDNA polymorphisms in Iraq, a crossroad of early human dispersal and of post-Neolithic migrations, Mol. Phylogenet. Evol. 28 (2003) 458–472. [7] C. Cinnioglu, R. King, T. Kivisild, E. Kalfoglu, S. Atasoy, G.L. Cavallery, A.S. Lillie, C.C. Roseman, et al. Excavating Y-chromosome haplotype strata in Anatolia, Hum. Genet. 114 (2004) 127–148. [8] J.J. Sanchez, M. Brion, W. Parson, A.J. Blanco-Verea, C. Boersting, M.V. Lareu, H. Niedersta¨tter, H. Oberacher, N. Morling, A. Carracedo, Duplications of the Y-chromosome specific loci P25 and 92R7 and forensic implications, Forensic Sci. Int. 140 (2004) 241–250. [9] L. Excoffier, P.E. Smouse, J.M. Quattro, Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data, Genetics 131 (1992) 479–491.

[10] A. Carracedo, A. Beckmann, A. Bengs, B. Brinkmann, A. Caglia, C. Capelli, P. Gill, L. Gusmao, et al. Results of a collaborative study of the EDNAP group regarding the reproducibility and robustness of the Y-chromosome STRs DYS19, DYS389 I and II, DYS390 and DYS393 in a PCR pentaplex format, Forensic Sci. Int. 119 (2001) 28–41. [11] L. Roewer, M. Krawczak, S. Willuweit, M. Nagy, C. Alves, A. Amorim, K. Anslinger, C. Augustin, et al. Online reference database of European Y-chromosomal short tandem repeat (STR) haplotypes, Forensic Sci. Int. 118 (2/3) (2001) 106– 113. [12] K. Bender, B. Stradmann-Bellinghausen, C. Rittner, P.M. Schneider, Comparative analysis of short tandem repeats and single nucleotide polymorphisms on the Y-chromosome in Germans, Chinese and Thais, Legal Med. 5 (Suppl. 1) (2003) S164–S168.